Traces

Every run: the agent's full trajectory, its submissions over the 20-hour budget, and the final score.

MEG Speech Decoding

task page →5 runs · indexed 2026-09-02
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1GPT-5.60.3119$41.2143.5M133k5h 23m2042
2GPT-5.60.3039$14.1019.5M85k2h 25m1372
3GPT-5.60.2311$153.75128.1M180k10h 45m29817
4GPT-5.60.1902$64.3567.7M163k7h 2m2794
5GPT-5.60.1862$104.8098.3M165k7h 53m3188